Networks that utilize small cells (e.g., Home eNodeB devices) are gaining momentum in the marketplace. Examples of small cells may include cells that are served by relatively lower power base stations, for example including relatively lower power evolved Node B (eNB) devices such as home eNBs (HeNBs), Relay Nodes (RNs), Remote Radio Heads (RRHs), and/or the like. The small cell base stations may have a relatively smaller coverage area as compared to a macro cell. Such small cells are often added to a network to increase capacity in areas with high levels of user and/or to provide additional coverage in areas not covered by the macro network—e.g., both outdoors and/or indoors. Small cells can also improve network performance and service quality by facilitating the offloading of traffic from the large macro-cells. Such heterogeneous networks with large macro-cells in combination with small cells can provide increased bitrates per unit area.
An offloading technique known as Selected IP Traffic Offload (SIPTO) may allow an operator to select a packet data network (PDN) gateway (PDN GW or P-GW) for one or more wireless transmit receive units (WTRUSs) that may take into account the location of the WTRU. The WTRU's PDN connection may be torn down and reestablished if the network realizes it may be advantageous to do so, for example, based on the location of the WTRU. Such techniques for reselecting a PDN GW that is closer to the actual location of the WTRU may facilitate more efficient routing of data within the core network, thereby more efficiently utilizing network resources. SIPTO may be used to enable local breakout of traffic from a small cell.
SIPTO may allow an operator to streamline an established PDN connection by reassigning a new P-GW that may be geographically closer to the current location of a WTRU. P-GW relocation may imply a change in IP address, and performing SIPTO may disrupt any ongoing services. It has been recommended that SIPTO should not be performed for WTRUs in a connected mode to avoid disrupting ongoing services. While this recommendation may represent an improvement compared to blindly performed SIPTO, it fails to address the issue of smooth P-GW relocation for WTRUs with long-lived and real-time IP flows, e.g. long conference calls, large file transfers, and the like.